Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
1, February 2015
INTERFERENCE MANAGEMENT IN
LTE DOWNLINK NETWORKS
Farhana Afroz1, Kumbesan Sandrasegaran2, H. Al Kim3
Faculty of Engineering and Information Technology,
University of Technology, Sydney, Australia
ABSTRACT
Two major challenges for evolving LTE (Long Term Evolution) networks are to achieve enhanced
system capacity and cell coverage compared with WCDMA (Wideband Code Division Multiple Access)
system. Effective utilization of radio resources as well as dense spectrum reuse are at the core to attain
these targets. However, dense frequency reuse may increase inter-cell interference, which in turn
severely limits the capacity of users in the system. Inter-cell interference can restrict overall system
performance in terms of throughput and spectral efficiency, especially for the users located at the cell
edge area. Hence, careful management of inter-cell interferences becomes crucial to improve LTE
system performance. In this paper, interference mitigation schemes for LTE downlink networks are
investigated.
KEYWORDS
LTE, OFDMA, interference avoidance, interference randomization, resource block, SINR, throughput
1.INTRODUCTION
The growing demand of providing ubiquitous broadband internet access on mobile networks
has imposed the need of developing OFDMA (Orthogonal Frequency Division Multiple
Access) based wireless cellular networks such as 4G networks. One of the major challenges
for evolving LTE networks is to increase network capacity [1]. Though dense frequency reuse
results in significant system capacity improvement, it also remarkably degrades the
performance of the system due to the increase in interference caused by adjacent cells [1, 2]. In
cellular mobile communication system, mainly two types of interference must be taken into
consideration such as intra-cell interference and inter-cell interference. In intra-cell
interference (shown in Fig. 1(a)), interfering mobile terminal is in the same cell. The spillover
transmission between adjacent channels within a cell results in intra-cell interference. In intercell interference (ICI) (shown in Fig. 1(b)), interfering mobile terminal is in adjacent cell. ICI
is caused by the use of the same frequency channel in neighbouring cells [3].
In LTE downlink (DL), OFDMA radio access technology is used where the subcarriers are
mutually orthogonal to each other, implying that there is no intra-cell interference. However,
ICI can limit system performance in terms of throughput and spectral efficiency, especially for
users located at the cell edge. So, careful management of inter-cell interference is very
important in LTE to improve system performance [3, 4].
To mitigate inter-cell interference (ICI), some strategies are used during the transmission or
after the reception of the signal. ICI mitigation techniques can be classified as interference
DOI : 10.5121/ijwmn.2015.7106
91
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
randomization, interference cancellation and interference avoidance. In the first strategy, some
cell specific scrambling, interleaving or spread spectrum techniques can be used to reduce
interference. The
interference is distributed randomly among all users such as using pseudorandom scrambling
after channel coding. Thus, the cell edge users will not always suffer strong ICI during the
entire transmission period. In interference cancellation, the interfering signal is regenerated
through signal processing and the estimated interfering signal is subtracted from the received
signal
(desired
signal
+
interference) [5].
[a]
[b]
Fig. 1: (a) Intra-cell interference [3] (b) Inter-cell interference [12]
92
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
(1)
Where,
is the SINR of UE m on RB n
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
which thereby increases interference, and when adjacent cells assign the same RB to their
UEs. Thus, the SINRs of cell centre users (CCU) located nearer to their serving eNB are better
as compared with the users located at the edges.
A range of power and frequency allocation strategies can be adopted for cell centre and cell
edge users to mitigate ICI. Though an increase in transmit power can improve the signal to
interference and noise ratio, it may significantly increase the overall interference of the
system. Thus for each UE, particularly cell edge user (CEU), an increase in users transmit
power imposes a conflict over the
overall system performance. So, various power allocation schemes are proposed to obtain a
trade-off between the achieved SINR and resulting interference so as the systems
performance can be improved [9]. Considering the third situation when ICI increases (when
the neighbor cells allocate the same RB to their users), various RB allocation schemes are
adopted as ICI mitigation schemes in which the objective of the schemes is to reduce ICI and
maintain the higher spectral efficiency concurrently. It is also observed from the equation (1)
that if the serving cell selects RBn to transmit data, the ICI can be reduced noticeably if the
adjacent cells do not assign the same RBn to their users i.e. when
. The allocation
granularity can be a resource block or a part of available bandwidth. The conception of
reserving specific portions of the bandwidth for CCU and CEU to avoid interference is named
as frequency reuse technique.
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
allocation or frequency resource usage (as shown in Fig. 3(a)). With this scheme, high system
capacity i.e. high peak data rate can be achieved. However, inter-cell interference, especially at
cell edges, is increased which in turn considerably limits the performance of cell-edge users.
Thus, the overall spectral efficiency degrades.
In reuse factor three (RF3), the total bandwidth is divided into three equal and orthogonal subbands and the sub-bands are allocated to cells in such a way that adjacent cells always deploy
different frequencies (shown in Fig. 3(b)). This scheme leads to lower inter-cell interference.
However, as each cell is using one third of total available bandwidth, there is a large capacity
loss.
(a)
(b)
Fig. 3: Conventional frequency planning (a) frequency reuse factor 1 (b) requency reuse factor 3 [6]
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
cell centre regions [13]. The basic concept of PFR is to put restrictions on a portion of the
resources so as some resources are not utilized by some user classes at all. In PFR scheme
(shown in Fig. 4), total available bandwidth is divided into four sub-bands. Cell centre UEs are
allocated in the frequency band using reuse factor of 1. Cell edge UEs are allocated in the
complementary frequency band using reuse factor of 3. This scheme is also known as FFR-FI
(FFR with full isolation) as the cell edge users are completely isolated. As PFR does not
employ the whole available bandwidth, it leads to lower cell throughput compared with RF1
scheme.
3.1.2.2 Soft Frequency Reuse (SFR)
One shortcoming of PFR scheme is it may under-utilize the available frequency resources
because of its restricted no-sharing policy. Soft frequency reuse scheme proposed in [14, 15],
was aimed to avoid the high inter-cell interference associated with reuse factor 1 configuration
while more flexibility is being provided to the PFR scheme. In SFR (shown in Fig. 5), each
cell uses the total available bandwidth. For each sector, cell edge users are allocated in the
fraction of bandwidth with highest power level and cell centre users are allocated with lower
power in the rest of the frequency band. RF1 is used in the cell centre region and FRF greater
than one is employed at the cell edge regions.
An enhancement of SFR scheme is known as SFFR (Soft Fractional Frequency Reuse). The
SFR and PFR techniques can enhance the throughput of the users at the cell edge region by
minimizing the inter-cell interference experienced by cell-edge users. However, these schemes
may result in lower cell throughput compared with conventional RF1 scheme. As PFR does
not employ the whole bandwidth available in the cell, it leads to lower cell throughput
compared with RF1 scheme. Furthermore, though SFR may improve overall system capacity
as compared with PFR (as SFR can utilize the whole available bandwidth), however SFR may
lead to lower overall system capacity than RF1 scheme. SFFR scheme was proposed to
improve the overall cell throughput. In soft FFR, the resources allocated to the users at the cell
edges of other cell are allowed to be used by the other cells users (inner) with some power
restriction (as shown in Fig. 6) [16].
96
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
Fig.6: Frequency planning and power allocation for SFFR scheme [16]
97
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
shortcomings of SFR scheme can be overcome by IFR scheme, it only provides better
performance under low traffic situation.
3.1.3.2 Enhanced Fractional Frequency Reuse (EFFR)
EEFR was proposed to further improve the IFR and SFR system performance. EFFR aims to
increase the system capacity particularly in overloaded traffic situation. As shown in Fig. 8,
like IFR scheme, EFFR scheme specifies three cell-types for directly adjoining cells, and for
each cell-type a portion of the total band namely Primary Segment are reserved. The Primary
Segments should be orthogonal. The rest sub-channels apart from the Primary Segment form
Secondary Segment. At the same time, the Primary Segment of a cell-type is a part of the
Secondary Segments of other 2 cell-types. All sub-channels in each cells Primary Segment
can be occupied by this cell at will, however only a portion of sub-channels in the Secondary
Segment can be occupied by this cell in inter-cell interference-aware manner. Each cells
Primary Segment is further divided into a RF3 (reuse-3) part and RF1 (reuse-1) part. The RF1
part can be reused by all cell-types, whereas RF3 part can be reused by other same cell-type.
The RF3 sub-channels are forbidden to be reused by directly adjoining cells, which in turn
reduces the co-channel interferences [17].
99
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
resource blocks to each eNB in such a way that capacity is maximized. Thus, each eNB need
to forward the channel state information received from each UE to the central control unit and
receive back the resource allocation information. But, it is hard task to do centralized
scheduling because of the stringent time needed to exchange scheduling information and the
large feedback information required by the users to transmit all the CSIs. Thats why, LTE-A
system has abolished the central control unit and relied on coordination among eNBs over the
X2 interface without any centralized coordination in a flat architecture [4]. Some examples of
this type of coordination based scheme can be found in [24-26].
3.2.2 Semi-Distributed Scheme
In semi-distributed schemes, coordination is usually implemented at two levels such as the
central controlling entity level and the eNBs level. Like centralized scheme, a central
controlling entity is deployed that controls a number of eNBs. However, in semi-distributed
approach, each super-frame of bulk resources are allocated to each eNB by the central entity
whereas in the centralized scheme the central entity allocates the channels directly to each user
on frame-basis. Hence, in semi-distributed scheme, each eNB is in charge to allocate channels
on the frame level to the users that are served. As the resource allocation task is distributed
between eNBs and central controlling entity, as a whole the computational load of the scheme
can be reduced. Some examples of semi-distributed schemes can be found in [27-30]. The
semi-distributed scheme can be employed for eICIC (enhanced Inter Cell Interference
Coordination) in HetNets (Heterogeneous Network) [6].
3.2.3 Coordinated-Distributed Scheme
In centralized and semi-distributed approaches, all the interference information on every
resource block need to be collected at the central control entity, and in reality, the volume of
this information required from eNBs to the central controller can be excessively large [29]. For
this reason, the rate of exchanging information between eNB and central entity must be
minimized, which results in degraded overall system performance. In coordinated-distributed
schemes, central entity doesnt need to perform the coordination and resources are allocated
only at the eNB level. But, still coordination between eNBs is required for exchanging CSI
reports to maintain global ICIC. For practical implementation, the coordinated-distributed
scheme is more efficient because it has distinct pros over the semi-distributed approach such
as minimizing time and signalling overhead due to the systematic communication between
eNBs and central controller, lessening the complexity of network infrastructure as central
controller is eliminated here. Some examples of coordinated-distributed schemes reported in
the literature can be found in [31-35]. Fig. 11 shows an example of coordinated-distributed
scheme.
101
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
Fig. 11: Dynamic fractional frequency reuse scheme [35] - an example of coordinateddistributed scheme where cell centre boundaries are changed dynamically based on cell load,
user behaviors, and interference situation from adjoin cells. As seen in figure, the cell centre
areas are of different size in different cell and cell 1 has high load while cell 3 has lowest load.
Consequently, the region of RF1 comes larger in cell 1 as compared with cell 3.
3.2.4 Autonomous-Distributed Scheme
In autonomous-distributed approaches, resource allocation is made only at the eNB level
without any usage of central controlling entity for coordination. This is the similarity of
autonomous-distributed schemes with coordinated-distributed schemes. On the other hand,
dissimilar to coordinated-distributed schemes, coordination between eNBs is not required for
autonomous-distributed schemes. For each eNB, channels are allocated by the eNB to its
corresponding UEs depending on the local information gathered from its UEs. In autonomousdistributed scheme, it is possible to place the RBs anywhere (in a distributed-fashion) as
required to enhance the system capacity, so the system acts as self-organizing system. In order
to maintain network-wide ICIC and good fairness with autonomous-distributed schemes, some
of the RBs of each eNB must be restricted by minimizing power level or not to use it at all
which in turn reduces the ICI on those RBs for adjacent cells. Since, there is no coordination
among eNBs in autonomous schemes, the RBs need to be restricted are selected on the basis of
SINR values of those RBs. Low SINR level specifies that a resource block is being utilized by
adjacent cells. When RBs are being restricted by eNB, a scheme requires to make a
compromise between the value of lessening the ICI in adjacent cells and the cost of utilizing
the spectrum available [7]. Some example of autonomous-distributed ICIC schemes can be
found in [36-40].
4. INTERFERENCE RANDOMIZATION
One of three ICI mitigation techniques is named as interference randomization. In interference
randomization policy, the users data are spread up over a distributed set of subcarriers so that
interference scenario can be randomized and frequency diversity gain can be achieved. In
interference randomization, in each cell the users data are sequentially allocated over a timefrequency chunks. When all the requested transmission are allocated, subcarriers permutation
is made in random manner so that each UEs transmission is arbitrarily spread up over the total
time-frequency grid. Fig. 12 shows the allocation of subcarriers in a given cell before and after
102
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
Fig. 12: Subcarrier allocation before (a) and after (b) random permutation [41]
5. CONCLUSION
Inter-cell interference coordination strategies can be considered as a scheduling scheme that
takes into account the situation of adjacent cells to reduce the impact of inter-cell interference
(IC) and improve the cell edge throughput. As a whole, the ICI avoidance schemes put
restrictions of using downlink resources such as time/frequency/transmit power. The
coordination of these restrictions helps to limit the generation of interference in a cellular
network. Therefore, at the receiving end within the network coverage, SINR can be improved,
which in turn provides an opportunity for getting increased data rate over the network
coverage area. In this paper, a survey of interference avoidance schemes employed to mitigate
the inter-cell interference problem occurred in downlink LTE system are studied. This paper
presents a review of three main types of interference avoidance schemes, namely, static, semistatic and dynamic schemes. Also the basic concept of interference randomization is discussed.
Our future work includes to review the existing interference cancellation schemes for LTE
network as well as to propose an interference mitigation scheme for downlink LTE system.
103
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
[25] S. Das, H. Viswanathan, and G. Rittenhouse, "Dynamic load balancing through coordinated
scheduling in packet data systems," in Proc. IEEE INFOCOM 2003, 2003, vol. 1, pp. 786-796.
[26] T. Bonald, S. Borst, and A. Proutiere, "Inter-Cell Scheduling in Wireless Data Networks," in Proc.
11th European Wireless Conference 2005, 2005, pp. 1-7.
[27] M. Rahman and H. Yanikomeroglu, Enhancing cell-edge performance: a downlink dynamic
interference avoidance scheme with inter-cell coordination," IEEE Transactions on Wireless
Communications, vol. 9, no. 4, pp.1414-1425, 2010.
[28] M. Rahman and H. Yanikomeroglu, "Multicell Downlink OFDM Subchannel Allocations Using
Dynamic Inter-cell Coordination," in Proc. IEEE Global Telecommunications Conf. GLOBECOM
'07, 2007, pp. 5220-5225.
[29] G. Li and H. Liu, Downlink Radio Resource Allocation for Multi-Cell OFDMA System," IEEE
Transactions on Wireless Communications, vol. 5, no. 12, pp. 3451-3459, 2006.
[30] M.C. Necker, A Novel Algorithm for Distributed Dynamic Interference Coordination in Cellular
Networks," in Proc. KiVS, pp. 233-238, 2011.
[31] A. Triki and L. Nuaymi, Intercell Interference Coordination Algorithms in OFDMA Wireless
Systems," in Proc. IEEE 73rd Vehicular Technology Conf. (VTC Spring), 2011, pp. 1-6.
[32] T. Q.S. Quek, Z. Lei, and S. Sun, Adaptive interference coordination in multi-cell OFDMA
systems," in IEEE 20th International Symposium on Personal, Indoor and Mobile Radio
Communications, 2009, pp. 2380-2384.
[33] K. Dong, H. Tian, X. Li, and Q. Sun, A Distributed Inter-Cell Interference Coordination Scheme
in Downlink Multicell OFDMA Systems," in Proc. 7th IEEE Consumer Communications and
Networking Conf. (CCNC), 2010, pp. 1-5.
[34] H. Kwon, W.-I. Lee, and B. Gi Lee, Low-Overhead Resource Allocation with Load Balancing in
Multi-cell OFDMA Systems," in IEEE 61st Vehicular Technology Conference, 2005, pp. 30633067.
[35] D. Kimura, Y. Harada, and H. Seki, De-Centralized Dynamic ICIC Using X2 Interfaces for
Downlink LTE Systems," in Proc. IEEE 73rd Vehicular Technology Conf. (VTC Spring), 2011,
pp. 1-5.
[36] A. L. Stolyar and H. Viswanathan, Self-Organizing Dynamic Fractional Frequency Reuse in
OFDMA Systems," in Proc. INFOCOM 2008. The 27th Conf. Computer Communications. IEEE,
2008, pp. 691-699.
[37] A. L. Stolyar and H. Viswanathan, Self-Organizing Dynamic Fractional Frequency Reuse for
Best-Effort Traffic through Distributed Inter-Cell Coordination," in Proc. IEEE INFOCOM 2009,
2009, pp. 1287-1295.
[38] S. Cicalo, V. Tralli, and A.I. Perez-Neira, Centralized vs Distributed Resource Allocation in
Multi-Cell OFDMA Systems," in IEEE 73rd Vehicular Technology Conference, 2011, pp. 1 - 6.
[39] S. Ko, H. Seo, H. Kwon, and B. G. Lee, Distributed Power Allocation for Efficient Inter-cell
Interference Management in Multi-cell OFDMA Systems," in 16th Asia-Pacific Conference on
Communications, 2010, pp. 243-248.
[40] Q. D. La, Y. H. Chew, and B.-H. Soong, \An Interference Minimization Game Theoretic
Subcarrier Allocation Algorithm for OFDMA-Based Distributed Systems," in IEEE Global
Telecommunications Conference, 2009, pp. 1-6.
[41] R. Bosisio and U. Spagnolini, Interference Coordination vs. Interference Randomization in
Multicell 3GPP LTE System, IEEE Wireless Communications and Networking Conference,
March April 2008, pp.824-829.
105
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 1, February 2015
Authors
Farhana Afroz obtained her MS degree and B.Sc. (Honours) degree from the
department of Applied Physics, Electronics and Communication Engineering of
University of Dhaka, Bangladesh. Her undergraduate and postgraduate-level
research works were in material science and digital signal processing respectively.
She is now pursuing Master of Engineering studies with Telecommunication
Engineering major in University of Technology, Sydney, Australia. She worked as a Lecturer for the
department of Electronics and Telecommunication Engineering in University of Liberal Arts
Bangladesh, during 2007-2010 time period. Her current research interest includes 4G mobile networks
and digital signal processing.
Dr Kumbesan Sandrasegaran is an Associate Professor at UTS and Centre for RealTime Information Networks (CRIN). He holds a PhD in Electrical Engineering from
McGill University (Canada) (1994), a Master of Science Degree in Telecommunication
Engineering from Essex University (1988) and a Bachelor of Science (Honours)
Degree in Electrical Engineering (First Class) (1985). His current research work
focuses on two main areas (a) radio resource management in mobile networks, (b) engineering of
remote monitoring systems for novel applications with industry through the use of embedded systems,
sensors and communications systems. He has published over 100 refereed publications and 20
consultancy reports spanning telecommunication and computing systems.
Haider Al Kim got his B.Sc. in Information and Communication Engineering from AlKhwarizmi Engineering College, University of Baghdad, Baghdad, Iraq in 2008. He
pursues his Master degree in Telecommunication Networks from University of
Technology Sydney (UTS), Sydney, Australia in 2014 under the supervision A. Prof.
Kumbesan Sandrasegaran. Working and research areas are Wireless Telecommunication,
Mobile Network, Network Management, Network Design and Implementation and Data Analysis and
Monitoring .He is senior network engineer with more than 5 years work experience in networks and
telecommunication industry at University of Kufa, Iraq. He is also a Cisco Certificate holder (ID:
CSCO11773718) and Cisco instructor at Al-Mansour College, Baghdad, Iraq in 2010-2011.
106